DE102021200182A1 - Calibration method for calibrating an angle sensor of an electrical machine, control unit and computer program - Google Patents

Abstract

The invention relates to a calibration method for calibrating an angle sensor (2) of an electrical machine (1), with which a rotor position of a rotor (3) of the electrical machine (1) is determined, the method having the following steps: - rotating the rotor (3 ) with an angular velocity (ω1) and detecting an angle (θS1) of the rotor position output by the angle sensor (2) in a state of the electrical machine (1) in which an actual angle (θt1) of the rotor position is known;- determining an offset angle (β1) from the output angle (θS1) and the actual angle (θt1) of the rotor position, the offset angle (β1) containing at least one speed-dependent angle component, depending on the angular velocity (ω1) and a sensor transit time (Td_real); characterized by- calculating the individual sensor transit time (Td_real) based on the determined offset angle (β1) and another offset angle (β2) at a different angle l speed (ω2) of the rotor (3) was determined; and- storing the calculated individual sensor running time (Td_real) to correct an angle (θS) output by the angle sensor (2), depending on the angular velocity of the rotor (3). of the procedure are set up.

Description

The invention relates to a calibration method for calibrating an angle sensor of an electrical machine, a control unit that is set up to carry out the calibration method, and a corresponding computer program for executing the calibration method on the control unit. Electrical machines are used in various applications in the prior art. For example, electric machines are used as drive units in hybrid vehicles or fully electric vehicles.

In the prior art, electrical machines are regulated by so-called field-oriented regulation, in which the position of the rotor of the electrical machine or the corresponding angle is taken into account as an input variable.

An angle sensor provided on the electrical machine detects the angle of the rotor position, which is taken into account as an input variable and which reflects the orientation of the magnetic poles of the rotor.

The angle sensor and the rotor are generally offset from one another due to the fact that they are not installed at the same time, so that there is a real offset angle between the angle output by the angle sensor and the actual position of the rotor.

According to the prior art, this offset angle can be determined when the rotor is stationary or rotating and can be used to correct the angle output by the angle sensor. Patent publication number DE 10 2008 001408 A1 describes the determination of the offset angle when the rotor is stationary or rotating.

In addition to the real offset angle, the angle output by the angle sensor has an additional error that is caused by the propagation time of the sensor signal from the angle sensor. The error increases as the rotor speed increases.

In the prior art, this additional error is counteracted in the event that the offset angle is determined when the rotor is at a standstill, in that the angle output by the angle sensor is based on a standard runtime - which can usually be found in the data sheet of the angle sensor - which is one of Corresponds to the average running time determined by the sensor manufacturer and applicable to the type of angle sensor.

In practice, this correction leads to deviations in the corrected angle if the individual and specifically used angle sensor has a running time that deviates from the average running time that can be found in the data sheet.

In the other case, in which the offset angle is determined during rotation of the rotor, the speed-dependent error is also taken into account in the offset angle determined and is therefore also corrected. However, this only applies to the specific speed at which the offset angle is determined.

For other speeds, the running time during operation of the electric machine is simply incorrectly compensated, which is why there are deviations. The deviations explained lead to an incorrect angle of the rotor that is taken into account in the field-oriented control and thus to less torque, less efficiency and greater instability of the electrical machine.

Against this background, the object of the invention is to create a calibration method--in particular for field-oriented regulation--of an electrical machine that allows a precise correction of a rotor angle output by an angle sensor of the electrical machine. It is at least the object of the invention to create an alternative calibration method to the prior art.

This task is solved by a calibration method according to patent claim 1, as well as a control device and computer program according to the further independent patent claims. Preferred embodiments are the subject of the dependent claims.

• - Rotieren des Rotors mit einer Winkelgeschwindigkeit und Erfassen eines Winkels der Rotorstellung mittels des Winkelsensors in einem Zustand der elektrischen Maschine, indem ein tatsächlicher Winkel der Rotorstellung bekannt ist;
• - Ermitteln eines Offsetwinkels aus dem erfassten Winkel und dem tatsächlichen Winkel der Rotorstellung, wobei der Offsetwinkel zumindest eine drehzahlabhängige Winkelkomponente beinhaltet, die sich aus der Winkelgeschwindigkeit und einer für den Winkelsensor individuellen Sensorlaufzeit ergibt;
• - Berechnen der individuellen Sensorlaufzeit auf Basis des ermittelten Offsetwinkels und eines weiteren Offsetwinkels, der bei einer gegenüber der Winkelgeschwindigkeit unterschiedlichen Winkelgeschwindigkeit des Rotors ermittelt wurde; und
• - Abspeichern der berechneten individuellen Sensorlaufzeit zum Korrigieren eines von dem Winkelsensor ausgegebenen Winkels der Rotorstellung, der bei einer bestimmten Winkelgeschwindigkeit des Rotors in einem Motorbetrieb der elektrischen Maschine von dem Winkelsensor ausgegeben wird.

A calibration method according to one aspect of the invention calibrates an angle sensor of an electrical machine, with which a rotor position of a rotor of the electrical machine is determined. The procedure has the following steps:

• - rotating the rotor at an angular velocity and detecting an angle of the rotor position by means of the angle sensor in a state of the electrical machine in which an actual angle of the rotor position is known;
• - Determining an offset angle from the detected angle and the actual angle of the rotor position, the offset angle containing at least one speed-dependent angle component which results from the angular velocity and an individual sensor running time for the angle sensor;
• - Calculation of the individual sensor running time based on the determined offset angle and a further offset angle, which was determined at an angular velocity of the rotor that differs from the angular velocity; and
• - Saving the calculated individual sensor propagation time to correct an angle of the rotor position output by the angle sensor, which is output by the angle sensor at a specific angular speed of the rotor when the electric machine is operating as a motor.

The rotor rotates in particular when the electric machine is in generator mode, with the rotor being rotated at the angular velocity by an external device. In the calibration method according to the invention, the external device is controlled, for example, by a user or by the control unit according to the invention itself.

The state of the electrical machine in which the actual angle of the rotor position is known can be determined, for example, from the induction voltages generated by the excited (self-excited or externally excited) rotor in the respective strands of the electrical machine.

The electrical machine whose angle sensor is calibrated using the calibration method according to the invention is, for example, a synchronous machine that is controlled by field-oriented control and in which the calibrated angle sensor supplies the angle indicating the rotor position as an input variable for the field-oriented control. The electric machine is preferably a drive assembly of a motor vehicle or a motorcycle.

In the calibration method according to the invention, the further offset angle is preferably detected in the same way, with the method step for determining the further offset angle differing only in that the angular velocity is different.

In addition to the speed-dependent angle component, the offset angle and the further offset angle can each contain an offset angle component which results from an orientation of an installation position of the angle sensor which is offset in relation to the actual angle of the rotor position. This is the case, for example, when the angle sensor is fitted or replaced without the actual alignment of the rotor being able to be verified.

und $${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{r}}+{\mathrm{\omega }}_2*{\text{T}}_{\text{d}\_\text{real}},$$

wobei β₁ = Offsetwinkel,
ω₁ = Winkelgeschwindigkeit bei Ermittlung des Offsetwinkel β₁,
θ_t1 = tatsächlicher Winkel der Rotorstellung bei Ermittlung des Offsetwinkels β₁,
θ_S1 = mittels (nicht-kalibriertem) Winkelsensor erfasster Winkel bei Ermittlung des Offsetwinkels β₁,
β_r = Offsetwinkelkomponente,
T_{d_real} = individuelle Sensorlaufzeit,
β₂ = weiterer Offsetwinkel,
ω₂ = Winkelgeschwindigkeit bei Ermittlung des weiteren Offsetwinkels β₂,
θ_t2 = tatsächlicher Winkel der Rotorstellung bei Ermittlung des weiteren Offsetwinkels β₂,
θ_S2 = mittels (nicht-kalibriertem) Winkelsensor erfasster Winkel bei Ermittlung des weiteren Offsetwinkels β₂,
die individuelle Sensorlaufzeit T_{d_real} in dem Berechnungsschritt berechnet werden durch: $${\text{T}}_{\text{d}\_\text{real}}=\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/\left({\mathrm{\omega }}_1\mathrm{-}{\mathrm{\omega }}_2\right).$$

When the offset angle component is present, the following relationships can be used: $${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_1*{\text{T}}_{\text{i.e}\_\text{real}},$$

and $${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_2*{\text{T}}_{\text{i.e}\_\text{real}},$$

where β ₁ = offset angle,
ω ₁ = angular velocity when determining the offset angle β ₁ ,
θ _t1 = actual angle of the rotor position when determining the offset angle β ₁ ,
θ _S1 = angle detected by means of a (non-calibrated) angle sensor when determining the offset angle β ₁ ,
β _r = offset angle component,
T _{d_real} = individual sensor running time,
β ₂ = further offset angle,
ω ₂ = angular velocity when determining the further offset angle β ₂ ,
θ _t2 = actual angle of the rotor position when determining the further offset angle β ₂ ,
θ _S2 = angle detected by means of (non-calibrated) angle sensor when determining the further offset angle β ₂ ,
the individual sensor transit time T _{d_real} can be calculated in the calculation step by: $${\text{T}}_{\text{i.e}\_\text{real}}=\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/\left({\mathrm{\omega }}_1\mathrm{-}{\mathrm{\omega }}_2\right).$$

The offset angle component can then be calculated from β _r =β ₂ -ω ₂ *T _{d_real} , or β ₁ -ω ₁ *T _{d_real} .

• - Bestromen mindestens einer Wicklung der elektrischen Maschine mit Gleichstrom derart, dass ein Magnetfeld mit bekannter Ausrichtung erzeugt wird; und
• - Erfassen eines von dem Winkelsensor ausgegebenen Winkels als den weiteren Offsetwinkel, nachdem sich der Rotor in Richtung des Magnetfeldes ausgerichtet hat und dadurch der tatsächliche Winkel der Rotorstellung bekannt ist.

The further offset angle can also be determined at an angular velocity of 0, i.e. when the rotor is at a standstill, with the further offset angle being determined in this case by:

• - Energizing at least one winding of the electrical machine with direct current in such a way that a magnetic field with a known orientation is generated; and
• - detecting an angle output by the angle sensor as the further offset angle after the rotor has aligned itself in the direction of the magnetic field and the actual angle of the rotor position is thereby known.

With this procedure, the further offset angle is determined on the basis of the relationship θ _t2 - θ _S2 = β ₂ = β _r , with the individual sensor running time then being calculated in the calculation step by: $${\text{T}}_{\text{i.e}\_\text{real}}=\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/{\mathrm{\omega }}_1.$$

$${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{r}}+{\mathrm{\omega }}_2*\left({\text{T}}_{\text{d}\_\text{real}}-{\text{T}}_0\right),$$

$${\text{T}}_{\text{d}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/\left({\mathrm{\omega }}_1\mathrm{-}{\mathrm{\omega }}_2\right);\text{und}$$

$${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{r}}+{\mathrm{\omega }}_1*\left({\text{T}}_{\text{d}\_\text{real}}-{\text{T}}_0\right)$$

$${\text{T}}_{\text{d}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/{\mathrm{\omega }}_1,\text{wobei}$$

bei Durchführung des erfindungsgemäßen Kalibrierverfahrens in diesem Fall die von dem Winkelsensor erfassten Winkel θ_S1 und θ_S2 eine Korrektur auf Basis der Standard-Sensorlaufzeit T₀ enthalten.The calibration method according to the invention can also be used in electrical machines with angle sensors that are already calibrated with a standard sensor runtime that is less precise than the individual sensor runtime. For example, the less accurate standard sensor runtime can be a be the average sensor running time specified by the sensor manufacturer, which the sensor manufacturer has determined for the type of angle sensor used. In this case, the standard sensor transit time T ₀ must be taken into account in the calibration method, with the above relationships then changing as follows for the calculation of the individual sensor transit time in the calculation step: $${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_1*\left({\text{T}}_{\text{i.e}\_\text{real}}-{\text{T}}_0\right),$$

$${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_2*\left({\text{T}}_{\text{i.e}\_\text{real}}-{\text{T}}_0\right),$$

$${\text{T}}_{\text{i.e}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/\left({\mathrm{\omega }}_1\mathrm{-}{\mathrm{\omega }}_2\right);\text{and}$$

$${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_1*\left({\text{T}}_{\text{i.e}\_\text{real}}-{\text{T}}_0\right)$$

$${\text{T}}_{\text{i.e}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/{\mathrm{\omega }}_1,\text{whereby}$$

when the calibration method according to the invention is carried out, in this case the angles θ _S1 and θ _S2 detected by the angle sensor contain a correction based on the standard sensor propagation time T ₀ .

Finally, the invention provides a computer program for a control unit for controlling an electrical machine, the computer program carrying out the calibration method explained when it is executed by the control unit.

The calibration method according to the invention, the corresponding control device and computer program have the effect that negative influences of the component variation of the angle sensor on the performance (torque and efficiency) and stability of the control are eliminated.

In vehicle applications, i. H. in motor vehicles or motorcycles, the invention allows the safety reserve to be reduced when performance is utilized. In addition, significantly cheaper angle sensors can be used, in which the sensor runtime does not have to adhere to narrow tolerances. Ultimately, the manufacture of angle sensors can also be simplified.

The following is a preferred embodiment of the invention with reference to the attached 1 explained.

1 shows schematically a field-oriented regulation of an electric machine, which is for example a drive unit of an automobile or a motorcycle and in which the method according to the invention for calibrating an angle output by an angle sensor is carried out.

1 shows an electric machine 1, which is an electric drive unit of an automobile or a motorcycle.

The electrical machine 1 is in particular an electrical synchronous machine that has three strands U, V, W. The strands U, V, W form a plurality of windings on a stator of the electrical machine 1, which are arranged around an interior space of the stator, preferably evenly distributed.

A rotor 3 is accommodated in the inner space of the stator and is rotatably supported there. An axis of rotation of the rotor 3 runs in 1 perpendicular to the drawing plane.

The three strands U, V, W are each connected to a bridge circuit 4, which converts a direct current supplied by a power supply, not shown, into alternating current. A high-voltage accumulator of the automobile or motorcycle takes over the energy supply, for example.

The alternating currents flowing in the three strands U, V, W generate a rotating magnetic field with a pole pair (p=1) in the interior, which is connected to the two poles of the in 1 shown rotor 3 interacts in such a way that the rotor 3 is set in rotation. In order to form the rotating magnetic field with a pair of poles, three windings are distributed around the interior of the stator, as is known. However, the invention is not limited to this. The electrical machine 1 can have a large number of pole pairs (p>=2), in which case the rotor also has corresponding 2*p poles.

The rotor 3 can be a rotor 3 that is self-excited by means of permanent magnet(s) or a rotor 3 that is externally excited by means of rotor windings. If the rotor 3 is separately excited, this is preferably done without brushes.

The AC currents flowing in the strands U, V, W are preferably regulated as part of a field-oriented regulation. the 1 schematically shows a corresponding block diagram.

The field-oriented control detects the alternating variables of the alternating currents flowing in the strands U, V, W and, based on this, carries out a Clarke transformation in a block 5 and a subsequent d/q transformation.

The former leads to the quantities I _α and I _β in the complex plane fixed to the stator with the real component α and the imaginary component jβ; the latter to the vectors Iq and Id in the d/q coordinate rotating with the rotor 3 data system. Vector Iq represents the torque and Vector Id represents the magnetic flux density of the rotating field. The quantities obtained by the transformations are in 1 shown schematically.

After determination of control deviations between the respective vectors and correspondingly desired reference variables (Refq and Refd), a respective PI control takes place at blocks 6. The reference variables reflect, for example, actuation of an accelerator pedal or handle by a driver.

Correspondingly controlled output variables of the PI controller 6 form input variables of the block 5'.

Block 5' performs an inverse d/q transform and an inverse Clarke transform, followed at block 7 by a space vector modulation. The output variables obtained from the block 7 are fed to the bridge circuit 4 in order to implement desired alternating currents in the strands U, V, W accordingly.

The previous explanations of the 1 illustrated field-oriented regulation are not to be understood as limiting. The aspect of the field-oriented control that is essential to the invention relates to a calibration method of a 1 shown angle sensor 2, which is explained below.

In general, the functions of the field-oriented control explained above and the calibration method according to the invention are implemented in a control unit, which executes the corresponding functions and method steps through the interaction of software and hardware.

In order to allow the d/q coordinate system to rotate correctly with the rotor 3 in the field-oriented control, it is necessary to know the exact rotor position of the rotor 3 . The exact rotor position is determined via the in 1 Angle sensor 2 shown, which outputs an angle indicating the rotor position. In addition, the angular velocity at which the d/q coordinate system or the rotor 3 rotates can be determined.

The angle sensor 2 can be a Hall sensor or an optical sensor, for example. Alternatively, the angle sensor 2 can be designed in such a way that it outputs sine or cosine signals that are phase-shifted with respect to one another, via which the angle can be determined.

The angular velocity results from the relationship ω=2*Pi*n, where n indicates a number of revolutions or a rotational frequency of the rotor 3 determined from a change in the output angle. Should the angle sensor 2, in particular in the embodiment with the mutually phase-shifted signals, contain a large number of sensor pole pairs (p _s ), i.e. run through several periods per complete revolution of the rotor 3, the stated relationship for determining the angular velocity is increased by the number of sensor pole pairs (ω=2*Pi*n*p _s ).

The output angle forms an input variable of the blocks 5 and 5' for carrying out the respective transformation and, as explained, should reflect the exact rotor position of the rotor.

In practice, however, this is difficult or hardly possible, because there is an offset angle (deviation) between the output angle and the actual rotor position, which is caused by the following circumstances.

On the one hand, angle sensor 2 has a sensor propagation time T _{d_real} , which angle sensor 2 requires in order to output the determined angle. In other words, the sensor transit time T _{d_real specifies} a delay time by which the output of the determined angle is delayed. The sensor propagation time T _{d_real} is not a fixed variable that is identical for all angle sensors 2 of the same specification, but an individual variable that varies for each individual angle sensor 2 . In reality, therefore, the offset angle contains an angular component that originates from the individual sensor transit time T _{d_real} and increases with increasing angular velocity.

On the other hand, the offset angle includes an offset angle component that results from an orientation of an installation position of the angle sensor 2 that is offset compared to the actual angle of the rotor position. In most cases, this real offset angle component results from the fact that at the time the angle sensor 2 is installed or replaced, the actual rotor position or actual angle of the rotor is not visible/known and no adjustment is possible.

θ_t = tatsächlicher Winkel der Rotorstellung, θ_S = von Winkelsensor ausgegebener Winkel, β_r = Offsetwinkelkomponente bedingt durch Einbaulage, ω = Winkelgeschwindigkeit und T_{d_real} = individuelle Sensorlaufzeit.In this respect, the following relationship results between the angle output by the angle sensor 2 and the actual angle of the rotor position: $${\mathrm{\theta }}_{\text{t}}={\mathrm{\theta }}_{\text{S}}+{\mathrm{\beta }}_{\text{right}}+\mathrm{\omega }*{\text{T}}_{\text{i.e}\_\text{real}}$$

θ _t = actual angle of the rotor position, θ _S = angle output by the angle sensor, β _r = offset angle component due to installation position, ω = Win kel speed and T _{d_real} = individual sensor transit time.

The above relationship shows that the offset angle between the actual angle θ _t of the rotor position and the angle θ _S output by angle sensor 2 is not constant, but contains the speed-dependent angle component that results from the individual sensor runtime T _{d_real} of the respective angle sensor 2 used and the instantaneous angular velocity ω of the rotor 3 results.

So that the speed-dependent angle component is correctly taken into account, the invention provides for calibrating the angle sensor 2 by determining the individual sensor running time T _{d_real} of the angle sensor 2 and then taking it into account in the field-oriented control when the electric machine 1 is operating as a motor.

For this purpose, in the calibration method according to the invention, the offset angle is detected in two different operating states or two method steps of the electric machine 1 and then used to calculate the individual sensor transit time T _{d_real} .

In the calibration method according to the invention, the at least two different operating states differ in particular in that the rotational speeds n or the angular speeds ω of the rotor 3 are different in the two operating states. What is the same, however, is that the electric machine 1 is preferably in generator mode in both operating states, i.e. the rotor 3 is preferably set in rotation by an external device. The order of the following method steps of the preferred calibration method is arbitrary.

• - Rotieren des Rotors 3 mit einer Winkelgeschwindigkeit ω₁ und Erfassen eines Winkels θ_S1 der Rotorstellung mittels des Winkelsensors 2 in einem Zustand der elektrischen Maschine 1, in dem ein tatsächlicher Winkel θ_t1 der Rotorstellung bekannt ist.

The electrical machine 1 is placed in one of the two different operating states and the following method step is carried out:

• - Rotating the rotor 3 with an angular velocity ω ₁ and detecting an angle θ _S1 of the rotor position by means of the angle sensor 2 in a state of the electrical machine 1 in which an actual angle θ _t1 of the rotor position is known.

The state of the electrical machine 1 in which the actual rotor angle θ _t1 is known can be derived, for example, from induction voltages which the excited rotor 3 generates in the strands or windings during generator operation. For example, the actual angle θ _t1 of the rotor position can be derived in a state in which the induction voltage of one of the three phases U, V, W has a zero crossing or the induction voltages of two phases are identical.

The offset angle β ₁ is then determined from the angle θ _S1 detected by the angle sensor 2 and the actual angle θ _t1 of the rotor position, with the following relationship applying: $${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_1*{\text{T}}_{\text{i.e}\_\text{real}}$$

The determined offset angle β ₁ contains on the one hand the speed-dependent angle component, which results from the angular velocity ω ₁ and the individual sensor running time T _{d_real} for the angle sensor 2, and on the other hand the offset angle component β _r , which results from the offset or offset angle of the rotor position compared to the actual angle .

• - Rotieren des Rotors 3 mit der unterschiedlichen Winkelgeschwindigkeit ω₂ und Erfassen eines Winkels θ_S2 der Rotorstellung mittels des Winkelsensors 2 in einem Zustand der elektrischen Maschine 1, in dem ein tatsächlicher Winkel θ_t2 der Rotorstellung bekannt ist.

After the offset angle β ₁ has been determined, the calibration method according to the invention proceeds to a second method step in which the electric machine 1 is switched to the second operating state by changing the rotational speed/angular speed and the method step explained is repeated as follows for the changed angular speed ω ₂ :

• - Rotating the rotor 3 with the different angular speed ω ₂ and detecting an angle θ _S2 of the rotor position by means of the angle sensor 2 in a state of the electric machine 1 in which an actual angle θ _t2 of the rotor position is known.

The state of the electrical machine 1 used in the second method step, in which the actual angle θ _t2 of the rotor position is known, can be identical to the state in the first method step or different - as long as the actual angle θ _t2 of the rotor position remains derivable .

The offset angle β ₂ can then be determined from the angle θ _S2 detected by the angle sensor 2 and the actual angle θ _t2 of the rotor position, where the following relationship applies: $${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_2*{\text{T}}_{\text{i.e}\_\text{real}}$$

The determined additional offset angle β ₂ in turn contains the speed-dependent angular component, which results from the changed angular velocity ω ₂ and the individual sensor running time T _{d_real} for the angle sensor 2, and on the other hand the offset angle component β _r , which consists of the angle θ _t2 of the rotor position offset or rotated installation position of the angle sensor 2 results.

und $${\mathrm{\beta }}_{\text{r}}={\mathrm{\beta }}_2-{\mathrm{\omega }}_2*{\text{T}}_{\text{d}\_\text{real}}.\text{oder}{\mathrm{\beta }}_1-{\mathrm{\omega }}_1*{\text{T}}_{\text{d}\_\text{real}}$$

After the offset angles β ₁ and β ₂ have been recorded, a calculation step follows in the calibration method according to the invention. Using the measured offset angles β ₁ and β ₂ and taking into account the above relationships (2) and (3), the calculation step results in the individual sensor transit time T _{d_real} and the offset angle component β _r from: $${\text{T}}_{\text{i.e}\_\text{real}}=\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/\left({\mathrm{\omega }}_1\mathrm{-}{\mathrm{\omega }}_2\right)$$

and $${\mathrm{\beta }}_{\text{right}}={\mathrm{\beta }}_2-{\mathrm{\omega }}_2*{\text{T}}_{\text{i.e}\_\text{real}}.\text{or}{\mathrm{\beta }}_1-{\mathrm{\omega }}_1*{\text{T}}_{\text{i.e}\_\text{real}}$$

The calculated variables are stored in a memory of the control unit. The calibration procedure ends at this point.

The electrical machine 1 can then switch to motor operation and start the field-oriented regulation already explained. The calculated and stored variables are used by the control unit during engine operation in order to correct the angle output by angle sensor 2 at the rotational speed/angular velocity actually present during engine operation in accordance with relationship (1) above.

In the calibration method explained, the offset angles are determined in operating states in which the conditions ω ₁ >0 and ω ₂ >0 apply. However, the invention is not limited to this.

In a preferred variant of the calibration method according to the invention, there is the possibility, in one of the two operating states or in one of the two method steps, that the state in which the actual angle of the rotor position is known is not in generator operation but when the rotor 3 is at a standstill—ergo at ω = 0 - to produce.

This takes place in that the control unit generates a current flow (direct current) in at least one strand or winding, which leads to the formation of a constant magnetic field with a known orientation. The excited rotor 3 aligns itself with the constant magnetic field, whereby the actual angle of the rotor position is also known due to the known orientation of the magnetic field.

If this procedure is selected, for example, in the second operating state or the second method step, the following relationship applies: $${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2\mathrm{=}{\mathrm{\beta }}_{\text{right}}$$

The further offset angle β ₂ determined corresponds exactly to the offset angle component β _r and, understandably, does not include any speed-dependent angle component.

In the calculation step, the individual sensor running time T _{d_real results} from the offset angle β ₁ determined in the first method step and the relationship (2) applicable to this method step, as well as the additional offset angle β ₂ determined in the second operating state and the corresponding relationship (6) above. . In this case, the calculation result is: $${\text{T}}_{\text{i.e}\_\text{real}}=\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/{\mathrm{\omega }}_1$$

The offset angle component β _r does not have to be calculated in this preferred variant of the calibration method according to the invention because it corresponds to the measured offset angle β ₂ in the second method step. The calculated variables are stored again and are used in the field-oriented control according to relation (1) to correct the deviation of the angle output by the angle sensor 2 .

In the calibration method according to the invention explained above and the corresponding preferred variant, the angle (θ _S1 or θ _S2 ) output by the angle sensor 2 corresponds to the raw, non-calibrated value of the angle sensor 2.

However, the method according to the invention can also be carried out when the angle sensor 2 has already been calibrated with a standard sensor transit time T ₀ which can be found, for example, in the data sheet for the angle sensor 2 .

$${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{r}}+{\mathrm{\omega }}_2*\left({\text{T}}_{\text{d}\_\text{real}}-{\text{T}}_0\right)$$

$${\text{T}}_{\text{d}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/\left({\mathrm{\omega }}_1\mathrm{-}{\mathrm{\omega }}_2\right)$$

$${\text{T}}_{\text{d}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/{\mathrm{\omega }}_1$$

In this case, the angle θ _S detected by the angle sensor 2 indicated in the above relationships (2)-(4), (7) corresponds to an angle corrected with the standard sensor transit time T ₀ . In order to use the calibration method in this case, the relationships (2)-(4), (7) must be supplemented by the standard sensor transit time T ₀ . The respective supplemented relationships are thus as follows: $${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_1*\left({\text{T}}_{\text{i.e}\_\text{real}}-{\text{T}}_0\right)$$

$${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_2*\left({\text{T}}_{\text{i.e}\_\text{real}}-{\text{T}}_0\right)$$

$${\text{T}}_{\text{i.e}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/\left({\mathrm{\omega }}_1\mathrm{-}{\mathrm{\omega }}_2\right)$$

$${\text{T}}_{\text{i.e}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/{\mathrm{\omega }}_1$$

Otherwise there are no differences to the procedure for non-calibrated angle sensor 2.

The aforementioned standard sensor runtime T ₀ is particularly preferably taken into account when the calibration method according to the invention is carried out on angle sensors that have already been calibrated with the standard sensor runtime as part of an update of the control unit.

The standard sensor runtime T ₀ stored in the control unit is preferably overwritten with the calculated individual sensor runtime T _{d_real} after the calibration method according to the invention has been successfully carried out, and the offset angle component β _r is also stored, so that the angle output by the angle sensor 2 is carried out in the subsequent engine operation , field-oriented control is corrected according to relation (1). The calibration method according to the invention can preferably also be used in electrical machines in which the angle sensor 2 and the rotor 3 are aligned exactly with one another and the offset angle component β _r is therefore equal to 0. In this case, the relationships are simplified, and the calibration method according to the invention only supplies the individual sensor transit time T _{d_real} , which is used in the subsequent engine operation and the corresponding field-oriented control.

The calibration method according to the invention and its variants are preferably implemented by a computer program that is installed on the named control unit or is installed as part of an update of the control unit.

QUOTES INCLUDED IN DESCRIPTION

This list of the documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• DE 102008001408 A1 [0005]

Claims (9)

Calibration method for calibrating an angle sensor (2) of an electrical machine (1), with which a rotor position of a rotor (3) of the electrical machine (1) is determined, the method having the following steps: - rotating the rotor (3) at an angular velocity (ω ₁ ) and detecting an angle (θ _S1 ) of the rotor position by means of the angle sensor (2) in a state of the electric machine (1) in which an actual angle (θ _t1 ) of the rotor position is known; - Determining an offset angle (β ₁ ) from the detected angle (θ _S1 ) and the actual angle (θ _t1 ) of the rotor position, the offset angle (β ₁ ) containing at least one speed-dependent angular component which results from the angular velocity (ω ₁ ) and a sensor running time (T _{d_real} ) that is individual to the angle sensor (2); characterized by - calculating the individual sensor running time (T _{d_real} ) on the basis of the determined offset angle (β ₁ ) and a further offset angle (β ₂ ), which occurs at an angular speed (ω ₂ ) of the rotor (3) that differs from the angular speed (ω ₁ ) was determined; and - storing the calculated individual sensor transit time (T _{d_real} ) to correct an angle (θ _S ) of the rotor position output by the angle sensor (2), which at a specific angular speed of the rotor (3) in motor operation of the electric machine (1) from the Angle sensor (2) is output. Calibration procedure according to Claim 1 , the further offset angle (β ₂ ) being determined by: - rotating the rotor (3) at the different angular velocity (ω ₂ ) and detecting an angle (θ _S2 ) of the rotor position by means of the angle sensor (2) in one state of the electrical machine (1) in which an actual angle (θ _t2 ) of rotor position is known; - Determining the further offset angle (β ₂ ) from the angle (θ _S2 ) detected at the different angular velocity (ω ₂ ) and the actual angle (θ _t2 ) of the rotor position, the further offset angle (β ₂ ) containing at least the speed-dependent angular component, which results from the different angular velocity (ω ₂ ) and the individual sensor running time (T _{d_real} ) for the angle sensor (2). Calibration procedure according to Claim 1 , wherein the further offset angle (β ₂ ) is determined by: - energizing at least one winding of the electrical machine (1) with direct current in such a way that a magnetic field with a known orientation is generated; and - detecting an angle (θ _S2 ) output by the angle sensor (2) as the further offset angle (β ₂ ) after the rotor (3) has aligned itself in the direction of the magnetic field and the actual angle (θ _t2 ) of the rotor position is thereby known is. Calibration procedure according to Claim 1 or 2 , wherein the offset angle (β ₁ ) and the further offset angle (β ₂ ) each contain an offset angle component (β _r ) which results from an orientation of an installation position of the angle sensor (2) which is offset in relation to the actual angle of the rotor position; and from relationships $${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_1*\left({\text{T}}_{\text{i.e}\_\text{real}}-{\text{T}}_0\right),$$

and $${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_2*\left({\text{T}}_{\text{i.e}\_\text{real}}-{\text{T}}_0\right),$$

where β ₁ = offset angle, ω ₁ = angular velocity when determining the offset angle β ₁ , θ _t1 = actual angle of the rotor position when determining the offset angle β ₁ , θ _S1 = angle detected by means of an angle sensor when determining the offset angle β ₁ , β _r = offset angle component, T _{d_real} = individual sensor transit time, T ₀ = standard sensor transit time, β ₂ = further offset angle, ω ₂ = angular velocity when determining the further offset angle β ₂ , θ _t2 = actual angle of the rotor position when determining the further offset angle β ₂ , θ _S2 = by means Angle sensor detected angle when determining the further offset angle β ₂ , is, the individual sensor running time T _{d_real} results from $${\text{T}}_{\text{i.e}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/\left({\mathrm{\omega }}_1\mathrm{-}{\mathrm{\omega }}_2\right).$$

Calibration procedure according to Claim 1 or 2 , wherein the offset angle (β ₁ ) and the further offset angle (β ₂ ) each contain an offset angle component (β _r ) which results from an orientation of an installation position of the angle sensor (2) which is offset in relation to the actual angle of the rotor position; and from relationships $${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_1*{\text{T}}_{\text{i.e}\_\text{real}},\text{and}$$

$${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_2*{\text{T}}_{\text{i.e}\_\text{real}},\text{whereby}$$

β ₁ = offset angle ω ₁ = angular velocity when determining the offset angle β ₁ , θ _t1 = actual angle of the rotor position when determining the offset angle β ₁ , θ _S1 = angle detected by means of an angle sensor when determining the offset angle β ₁ , β _r = offset angle component, T _{d_real} = individual sensor transit time, β ₂ = further offset angle, ω ₂ = angular velocity at Determination of the additional offset angle β ₂ , θ _t2 = actual angle of the rotor position when determining the additional offset angle β ₂ , θ _S2 = angle detected by means of an angle sensor when determining the additional offset angle β ₂ , actual, the individual sensor running time T _{d_real} results from $${\text{T}}_{\text{i.e}\_\text{real}}=\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/\left({\mathrm{\omega }}_1\mathrm{-}{\mathrm{\omega }}_2\right).$$

Calibration procedure according to patent claim 3 , wherein the offset angle (β ₁ ) and the further offset angle (β ₂ ) each contain an offset angle component (β _r ), which results from an orientation of an installation position of the angle sensor (2) that is offset relative to the actual angle of the rotor position; and from relationships $${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_1*\left({\text{T}}_{\text{i.e}\_\text{real}}-{\text{T}}_0\right),\text{and}$$

$${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{right}},\text{whereby}$$

β ₁ = offset angle ω ₁ = angular velocity when determining the offset angle β ₁ , θ _t1 = actual angle of the rotor position when determining the offset angle β ₁ , θ _S1 = angle detected by means of an angle sensor when determining the offset angle β ₁ , β _r = offset angle component, T _{d_real} = individual sensor transit time, T ₀ = standard sensor transit time, β ₂ = further offset angle, θ _t2 = actual angle of the rotor position when determining the further offset angle β ₂ , θ _S2 = angle detected by means of an angle sensor when determining the further offset angle β ₂ , is, itself the individual sensor running time T _{d_real} results from $${\text{T}}_{\text{i.e}\_\text{real}}={\text{T}}_0+\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/{\mathrm{\omega }}_1.$$

Calibration procedure according to patent claim 3 , wherein the offset angle (β ₁ ) and the further offset angle (β ₂ ) each contain an offset angle component (β _r ) which results from an orientation of an installation position of the angle sensor (2) which is offset in relation to the actual angle of the rotor position; and from relationships $${\mathrm{\theta }}_{\text{t}1}-{\mathrm{\theta }}_{\text{S}1}={\mathrm{\beta }}_1={\mathrm{\beta }}_{\text{right}}+{\mathrm{\omega }}_1*{\text{T}}_{\text{i.e}\_\text{real}},\text{and}$$

$${\mathrm{\theta }}_{\text{t}2}-{\mathrm{\theta }}_{\text{S}2}={\mathrm{\beta }}_2={\mathrm{\beta }}_{\text{right}},\text{whereby}$$

β ₁ = offset angle ω ₁ = angular velocity when determining the offset angle β ₁ , θ _t1 = actual angle of the rotor position when determining the offset angle β ₁ , θ _S1 = angle detected by means of an angle sensor when determining the offset angle β ₁ , β _r = offset angle component, T _{d_real} = individual sensor transit time, T ₀ = standard sensor transit time, β ₂ = further offset angle, θ _t2 = actual angle of the rotor position when determining the further offset angle β ₂ , θ _S2 = angle detected by means of an angle sensor when determining the further offset angle β ₂ , is, itself the individual sensor transit time T _{d_real} results from $${\text{T}}_{\text{i.e}\_\text{real}}=\left({\mathrm{\beta }}_1\mathrm{-}{\mathrm{\beta }}_2\right)/{\mathrm{\omega }}_1.$$

Control unit for controlling an electric machine (1), wherein the control unit is set up, - current alternating values of currents of individual strands of the electric machine (1), an angle of a rotor position of a rotor (3) of the electric machine (1) output by an angle sensor (2). ) and to detect an angular velocity (ω) of the rotor (3), - a calibration method according to one of patent claims 1 until 6 - to correct the output angle, taking into account the individual sensor transit time (T _{d_real} ), and - to carry out a field-oriented control for controlling the electrical machine (1) on the basis of the detected alternating current variables and the corrected angle. Computer program for a control unit for controlling an electrical machine (1), wherein the computer program, when executed by the control unit, uses a calibration method according to one of patent claims 1 until 7 performs.

Images